Abstract 3673Poster Board III-609 BackgroundThe treatment of HIV associated BL remains controversial with some concerned about the potential toxicity of the dose-intensive regimens typically used as treatment for BL in HIV seronegative patients. However, less intensive regimens have a high relapse rate. We modified the CODOX-M/IVAC regimen with the goal of preserving efficacy while improving tolerability, particularly treatment related mortality (TRM). The primary object was to improve the 1 year overall survival (OS) from 65% based on historical reports to 85%. An interim analysis was planned to exclude excessive TRM. MethodsMajor modifications of the original US NCI regimen include the addition of rituximab, reduction in cyclophosphamide to 800 mg/m2 × 2 days, vincristine 2 mg cap, methotrexate 3000 mg/m2, reduction of lumbar punctures using combination chemotherapy and infusion of the IVAC portion for the high risk patients. Antibiotic prophylaxis and growth factor support were specified as grade IV hematopoietic toxicities were reported in 100% of patients in the original regimen. HAART therapy was continued at the discretion of the local investigator. ResultsAs of June 2009, 22 of 31 planned patients have accrued and the study is ongoing. Baseline characteristics include: Classical Burkitt's, 95%; Low/High Risk, 9/91%; Median (range) Age 40 (19 – 55); CD4 count 290 (0 – 1260), CD4 <100, 5 ( 27%); HIV viral load 15,600 (48 – 715,881).Treatment StatusN (%)Treatment Completed per protocol13 (65%)Early termination due to adverse event*3 (14%)Early termination due to disease progression (central nervous system)1 (5%)Early termination due to patient withdrawal2 (9%)Early termination – counts did not recover within time frame to begin cycle 41 (5%)Treatment termination-reason not given1 (5%)Treatment ongoing2 (9%)*Terminations due to adverse events: 1 pt. due to grade 4 thrombocytopenia and grade 3 infection possibly related to doxorubicin and cyclophosphamide; 1 pt. due to grade 3 left hemiparesis possibly related to cytarabine and vincristine; 1 pt. due to grade 3 confusion unrelated to treatment.# of pts with Grade 3/4Treatment related toxicity: 15 with any, 13 hematologic, 7 with infection including 5 febrile neutropenia, 6 metabolic including 1 tumor lysis syndrome, 4 neurologic, 2 thrombotic and 1 each coagulation, GI or pain. Notably, only 2 patients had grade 1/2 stomatitis/mucositis and none had grade 3/4.Two patients died after treatment. Patient #1 died 3 months after completing treatment due to a fungal infection. Patient #4 died 5 months after completing treatment due to primary malignancy.<>Overall Survival<>N22<>Median (months)17.0<>One-year overall survival (95% CI)85.7% (60%, 100%) ConclusionsThe early stopping rule was not invoked for this dose modified, but intensive trial of HIV associated Burkitt's. In fact we had had no TRM. Notably, despite concerns raised in AMC 010, our prior study of DLBCL, rituximab did not appear to increase toxicity. Only 4 patients withdrew due to adverse events and 1 did not complete the protocol. The latter patient remains in remission, but has less than 1 year of follow up. Grade 3/4 toxicities were markedly reduced. These results compare favorably with 2 studies that excluded HIV + pts. Magrath (1995) originally reported 100% grade 4 hematologic and 20% mucositis in 39 adults, 33 children (92% 2 yr EFS). MRC/NCRI LY10 trial (Mead 2008) also reduced methotrexate(3gr/m2), but reported 9% treatment related deaths (64% 2 yr OS). Treatment efficacy thus far appears preserved in AMC 048 with our modifications which reduced TRM and morbidity. Correlative studies to improve our understanding of HIV-BL are pending. Accrual is ongoing. With ongoing enrollment and follow-up, it is possible the primary objective of 1 year OS of 85% may be achieved. Disclosures:No relevant conflicts of interest to declare.
Read full abstract